wassname/catalyst
1
0
Fork 0
mirror of https://github.com/wassname/catalyst.git synced 2026-06-29 13:47:54 +08:00
Code Issues Packages Projects Releases Wiki Activity

ENH: add dividend type column

Browse Source 
BUG: add back constant
This commit is contained in:
Maya Tydykov
2016-05-12 16:08:54 -04:00
parent 8a3b82c536
commit 3d0764a50c
6 changed files with 137 additions and 26 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tests/pipeline/test_dividends.py
+71 -17
View File
@@ -8,23 +8,28 @@ from six import iteritems
from zipline.pipeline.common import (
ANNOUNCEMENT_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
DAYS_SINCE_PREV_DIVIDEND_ANNOUNCEMENT,
DAYS_SINCE_PREV_EX_DATE,
DAYS_TO_NEXT_EX_DATE,
DIVIDEND_TYPE_FIELD_NAME,
EX_DATE_FIELD_NAME,
NEXT_AMOUNT,
NEXT_CURRENCY_TYPE,
NEXT_DIVIDEND_TYPE,
NEXT_EX_DATE,
NEXT_PAY_DATE,
PAY_DATE_FIELD_NAME,
PREVIOUS_AMOUNT,
PREVIOUS_ANNOUNCEMENT,
PREVIOUS_CURRENCY_TYPE,
PREVIOUS_DIVIDEND_TYPE,
PREVIOUS_EX_DATE,
PREVIOUS_PAY_DATE,
PREVIOUS_AMOUNT,
SID_FIELD_NAME,
TS_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
EX_DATE_FIELD_NAME,
PAY_DATE_FIELD_NAME,
CURRENCY_FIELD_NAME, NEXT_CURRENCY_TYPE)
)
from zipline.pipeline.data.dividends import (
DividendsByAnnouncementDate,
DividendsByExDate,
@@ -63,7 +68,8 @@ dividends_cases = [
PAY_DATE_FIELD_NAME: pd.to_datetime(['2014-01-15', '2014-01-20']),
TS_FIELD_NAME: pd.to_datetime(['2014-01-05', '2014-01-10']),
ANNOUNCEMENT_FIELD_NAME: pd.to_datetime(['2014-01-04', '2014-01-09']),
CURRENCY_FIELD_NAME: ["$", "EUR"]
CURRENCY_FIELD_NAME: ["$", "EUR"],
DIVIDEND_TYPE_FIELD_NAME: ["Stock", "Mixed"]
}),
# K1--K2--A2--A1.
pd.DataFrame({
@@ -72,7 +78,8 @@ dividends_cases = [
PAY_DATE_FIELD_NAME: pd.to_datetime(['2014-01-20', '2014-01-15']),
TS_FIELD_NAME: pd.to_datetime(['2014-01-05', '2014-01-10']),
ANNOUNCEMENT_FIELD_NAME: pd.to_datetime(['2014-01-04', '2014-01-09']),
CURRENCY_FIELD_NAME: ["EUR", "$"]
CURRENCY_FIELD_NAME: ["EUR", "$"],
DIVIDEND_TYPE_FIELD_NAME: ["Mixed", "Stock"]
}),
# K1--A1--K2--A2.
pd.DataFrame({
@@ -81,7 +88,8 @@ dividends_cases = [
PAY_DATE_FIELD_NAME: pd.to_datetime(['2014-01-10', '2014-01-20']),
TS_FIELD_NAME: pd.to_datetime(['2014-01-05', '2014-01-15']),
ANNOUNCEMENT_FIELD_NAME: pd.to_datetime(['2014-01-04', '2014-01-14']),
CURRENCY_FIELD_NAME: ["$", "EUR"]
CURRENCY_FIELD_NAME: ["$", "EUR"],
DIVIDEND_TYPE_FIELD_NAME: ["Stock", "Mixed"]
}),
# K1 == K2.
pd.DataFrame({
@@ -90,7 +98,8 @@ dividends_cases = [
PAY_DATE_FIELD_NAME: pd.to_datetime(['2014-01-10', '2014-01-15']),
TS_FIELD_NAME: pd.to_datetime(['2014-01-05'] * 2),
ANNOUNCEMENT_FIELD_NAME: pd.to_datetime(['2014-01-04', '2014-01-04']),
CURRENCY_FIELD_NAME: ["$", "EUR"]
CURRENCY_FIELD_NAME: ["$", "EUR"],
DIVIDEND_TYPE_FIELD_NAME: ["Stock", "Mixed"]
}),
pd.DataFrame(
columns=[CASH_AMOUNT_FIELD_NAME,
@@ -98,7 +107,8 @@ dividends_cases = [
PAY_DATE_FIELD_NAME,
TS_FIELD_NAME,
ANNOUNCEMENT_FIELD_NAME,
CURRENCY_FIELD_NAME],
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME],
dtype='datetime64[ns]'
),
]
@@ -174,6 +184,16 @@ next_currency_types = [[None, "$", "EUR", None],
[None, "$", None, "EUR", None],
[None, "$", "EUR", None]]
prev_dividend_types = [[None, "Stock", "Mixed"],
[None, "Stock", "Mixed"],
[None, "Stock", "Mixed"],
[None, "Stock", "Mixed"]]
next_dividend_types = [[None, "Stock", "Mixed", None],
[None, "Mixed", "Stock", "Mixed", None],
[None, "Stock", None, "Mixed", None],
[None, "Stock", "Mixed", None]]
class DividendsByAnnouncementDateTestCase(WithPipelineEventDataLoader,
ZiplineTestCase):
@@ -188,6 +208,8 @@ class DividendsByAnnouncementDateTestCase(WithPipelineEventDataLoader,
BusinessDaysSinceDividendAnnouncement(),
PREVIOUS_CURRENCY_TYPE:
DividendsByAnnouncementDate.previous_currency.latest,
PREVIOUS_DIVIDEND_TYPE:
DividendsByAnnouncementDate.previous_type.latest,
}
@classmethod
@@ -223,8 +245,10 @@ class DividendsByAnnouncementDateTestCase(WithPipelineEventDataLoader,
['NaT', '2014-01-04', '2014-01-14'],
['NaT', '2014-01-04']]
amounts = [['NaN', 1, 15], ['NaN', 7, 13], ['NaN', 3, 1], ['NaN', 23]]
currency_types = [[None, "$", "EUR"], [None, "$", "EUR"],
[None, "$", "EUR"]]
currency_types = [[None, "$", "EUR"], [None, "EUR", "$"],
[None, "$", "EUR"], [None, "EUR"]]
dividend_types = [[None, "Stock", "Mixed"], [None, "Mixed", "Stock"],
[None, "Stock", "Mixed"], [None, "Mixed"]]
cols = {
PREVIOUS_ANNOUNCEMENT: self.get_sids_to_frames(
zip_with_dates, announcement_dates, date_intervals, dates,
@@ -237,6 +261,10 @@ class DividendsByAnnouncementDateTestCase(WithPipelineEventDataLoader,
zip_with_strs, currency_types, date_intervals, dates,
'category', None
),
PREVIOUS_DIVIDEND_TYPE: self.get_sids_to_frames(
zip_with_strs, dividend_types, date_intervals, dates,
'category', None
),
}
cols[
@@ -260,7 +288,9 @@ class BlazeDividendsByAnnouncementDateTestCase(
ANNOUNCEMENT_FIELD_NAME: df[ANNOUNCEMENT_FIELD_NAME],
TS_FIELD_NAME: df[TS_FIELD_NAME],
SID_FIELD_NAME: sid,
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME]
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME],
CURRENCY_FIELD_NAME: df[CURRENCY_FIELD_NAME],
DIVIDEND_TYPE_FIELD_NAME: df[DIVIDEND_TYPE_FIELD_NAME],
})
for sid, df in iteritems(mapping)
).reset_index(drop=True)),)
@@ -290,6 +320,8 @@ class DividendsByExDateTestCase(WithPipelineEventDataLoader, ZiplineTestCase):
PREVIOUS_AMOUNT: DividendsByExDate.previous_amount.latest,
PREVIOUS_CURRENCY_TYPE: DividendsByExDate.previous_currency.latest,
NEXT_CURRENCY_TYPE: DividendsByExDate.next_currency.latest,
PREVIOUS_DIVIDEND_TYPE: DividendsByExDate.previous_type.latest,
NEXT_DIVIDEND_TYPE: DividendsByExDate.next_type.latest,
DAYS_TO_NEXT_EX_DATE: BusinessDaysUntilNextExDate(),
DAYS_SINCE_PREV_EX_DATE: BusinessDaysSincePreviousExDate()
}
@@ -331,7 +363,15 @@ class DividendsByExDateTestCase(WithPipelineEventDataLoader, ZiplineTestCase):
NEXT_CURRENCY_TYPE: self.get_sids_to_frames(
zip_with_strs, next_currency_types, next_date_intervals, dates,
'category', None
)
),
PREVIOUS_DIVIDEND_TYPE: self.get_sids_to_frames(
zip_with_strs, prev_dividend_types, prev_date_intervals, dates,
'category', None
),
NEXT_DIVIDEND_TYPE: self.get_sids_to_frames(
zip_with_strs, next_dividend_types, next_date_intervals, dates,
'category', None
),
}
cols[DAYS_TO_NEXT_EX_DATE] = self._compute_busday_offsets(
@@ -356,7 +396,9 @@ class BlazeDividendsByExDateLoaderTestCase(DividendsByExDateTestCase):
EX_DATE_FIELD_NAME: df[EX_DATE_FIELD_NAME],
TS_FIELD_NAME: df[TS_FIELD_NAME],
SID_FIELD_NAME: sid,
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME]
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME],
CURRENCY_FIELD_NAME: df[CURRENCY_FIELD_NAME],
DIVIDEND_TYPE_FIELD_NAME: df[DIVIDEND_TYPE_FIELD_NAME],
})
for sid, df in iteritems(mapping)
).reset_index(drop=True)),)
@@ -386,6 +428,8 @@ class DividendsByPayDateTestCase(WithPipelineEventDataLoader, ZiplineTestCase):
PREVIOUS_AMOUNT: DividendsByPayDate.previous_amount.latest,
PREVIOUS_CURRENCY_TYPE: DividendsByPayDate.previous_currency.latest,
NEXT_CURRENCY_TYPE: DividendsByPayDate.next_currency.latest,
PREVIOUS_DIVIDEND_TYPE: DividendsByPayDate.previous_type.latest,
NEXT_DIVIDEND_TYPE: DividendsByPayDate.next_type.latest,
}
@classmethod
@@ -425,7 +469,15 @@ class DividendsByPayDateTestCase(WithPipelineEventDataLoader, ZiplineTestCase):
NEXT_CURRENCY_TYPE: self.get_sids_to_frames(
zip_with_strs, next_currency_types, next_date_intervals, dates,
'category', None
)
),
PREVIOUS_DIVIDEND_TYPE: self.get_sids_to_frames(
zip_with_strs, prev_dividend_types, prev_date_intervals, dates,
'category', None
),
NEXT_DIVIDEND_TYPE: self.get_sids_to_frames(
zip_with_strs, next_dividend_types, next_date_intervals, dates,
'category', None
),
}
@@ -442,7 +494,9 @@ class BlazeDividendsByPayDateLoaderTestCase(DividendsByPayDateTestCase):
PAY_DATE_FIELD_NAME: df[PAY_DATE_FIELD_NAME],
TS_FIELD_NAME: df[TS_FIELD_NAME],
SID_FIELD_NAME: sid,
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME]
CASH_AMOUNT_FIELD_NAME: df[CASH_AMOUNT_FIELD_NAME],
CURRENCY_FIELD_NAME: df[CURRENCY_FIELD_NAME],
DIVIDEND_TYPE_FIELD_NAME: df[DIVIDEND_TYPE_FIELD_NAME],
})
for sid, df in iteritems(mapping)
).reset_index(drop=True)),)
tests/pipeline/test_earnings.py
+2 -1
View File
@@ -55,7 +55,8 @@ class EarningsCalendarLoaderTestCase(WithNextAndPreviousEventDataLoader,
'datetime64[ns]', 'NaN'
),
NEXT_ANNOUNCEMENT: self.get_expected_next_event_dates(
dates, 'datetime64[ns]', 'NaN'),
dates, 'datetime64[ns]', 'NaN'
),
}
cols[DAYS_TO_NEXT] = self._compute_busday_offsets(
cols[NEXT_ANNOUNCEMENT]
zipline/pipeline/common.py
+5 -3
View File
@@ -22,6 +22,7 @@ DAYS_SINCE_PREV_EX_DATE = 'days_since_prev_ex_date'
DAYS_TO_NEXT = 'days_to_next'
DAYS_TO_NEXT_EX_DATE = 'days_to_next_ex_date'
DISCLOSURE_DATE = 'disclosure_date'
DIVIDEND_TYPE_FIELD_NAME = 'dividend_type'
EX_DATE_FIELD_NAME = 'ex_date'
FISCAL_QUARTER_FIELD_NAME = 'fiscal_quarter'
FISCAL_YEAR_FIELD_NAME = 'fiscal_year'
@@ -32,6 +33,7 @@ NEXT_AMOUNT = 'next_amount'
NEXT_ANNOUNCEMENT = 'next_announcement'
NEXT_COUNT = 'next_count'
NEXT_CURRENCY_TYPE = 'next_currency_type'
NEXT_DIVIDEND_TYPE = 'next_dividend_type'
NEXT_EX_DATE = 'next_ex_date'
NEXT_FISCAL_QUARTER = 'next_fiscal_quarter'
NEXT_FISCAL_YEAR = 'next_fiscal_year'
@@ -51,12 +53,12 @@ PREVIOUS_BUYBACK_AMOUNT = 'previous_value'
PREVIOUS_BUYBACK_ANNOUNCEMENT = 'previous_buyback_announcement'
PREVIOUS_BUYBACK_CASH = 'previous_buyback_cash'
PREVIOUS_BUYBACK_SHARE_COUNT = 'previous_buyback_share_count'
PREVIOUS_DISCLOSURE_DATE = 'previous_disclosure_date'
PREVIOUS_COUNT = 'previous_count'
PREVIOUS_BUYBACK_TYPE = 'previous_buyback_type'
PREVIOUS_VALUE = 'previous_value'
PREVIOUS_VALUE_TYPE = 'previous_value_type'
PREVIOUS_BUYBACK_UNIT = 'previous_buyback_unit'
PREVIOUS_CURRENCY_TYPE = 'previous_currency_type'
PREVIOUS_DISCLOSURE_DATE = 'previous_disclosure_date'
PREVIOUS_DIVIDEND_TYPE = 'previous_dividend_type'
PREVIOUS_EX_DATE = 'previous_ex_date'
PREVIOUS_FISCAL_QUARTER = 'previous_fiscal_quarter'
PREVIOUS_FISCAL_YEAR = 'previous_fiscal_year'
zipline/pipeline/data/dividends.py
+5
View File
@@ -17,6 +17,8 @@ class DividendsByExDate(DataSet):
previous_amount = Column(float64_dtype)
next_currency = Column(categorical_dtype)
previous_currency = Column(categorical_dtype)
previous_type = Column(categorical_dtype)
next_type = Column(categorical_dtype)
class DividendsByPayDate(DataSet):
@@ -26,9 +28,12 @@ class DividendsByPayDate(DataSet):
previous_amount = Column(float64_dtype)
next_currency = Column(categorical_dtype)
previous_currency = Column(categorical_dtype)
previous_type = Column(categorical_dtype)
next_type = Column(categorical_dtype)
class DividendsByAnnouncementDate(DataSet):
previous_announcement_date = Column(datetime64ns_dtype)
previous_amount = Column(float64_dtype)
previous_currency = Column(categorical_dtype)
previous_type = Column(categorical_dtype)
zipline/pipeline/loaders/blaze/dividends.py
+12 -2
View File
@@ -2,6 +2,7 @@ from zipline.pipeline.common import (
ANNOUNCEMENT_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME,
EX_DATE_FIELD_NAME,
PAY_DATE_FIELD_NAME,
SID_FIELD_NAME,
@@ -49,6 +50,7 @@ class BlazeDividendsByAnnouncementDateLoader(BlazeEventsLoader):
{CASH_AMOUNT_FIELD_NAME}: ?float64,
{ANNOUNCEMENT_FIELD_NAME}: ?datetime,
{CURRENCY_FIELD_NAME}: ?string,
{DIVIDEND_TYPE_FIELD_NAME}: ?string,
}}
Where each row of the table is a record including the sid to identify the
@@ -65,6 +67,7 @@ class BlazeDividendsByAnnouncementDateLoader(BlazeEventsLoader):
CASH_AMOUNT_FIELD_NAME=CASH_AMOUNT_FIELD_NAME,
ANNOUNCEMENT_FIELD_NAME=ANNOUNCEMENT_FIELD_NAME,
CURRENCY_FIELD_NAME=CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME=DIVIDEND_TYPE_FIELD_NAME,
)
_expected_fields = frozenset({
@@ -72,7 +75,8 @@ class BlazeDividendsByAnnouncementDateLoader(BlazeEventsLoader):
SID_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
ANNOUNCEMENT_FIELD_NAME
ANNOUNCEMENT_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME
})
concrete_loader = DividendsByAnnouncementDateLoader
@@ -108,6 +112,7 @@ class BlazeDividendsByExDateLoader(BlazeEventsLoader):
{EX_DATE_FIELD_NAME}: ?datetime,
{CASH_AMOUNT_FIELD_NAME}: ?datetime,
{CURRENCY_FIELD_NAME}: ?string,
{DIVIDEND_TYPE_FIELD_NAME}: ?string,
}}
Where each row of the table is a record including the sid to identify the
@@ -124,6 +129,7 @@ class BlazeDividendsByExDateLoader(BlazeEventsLoader):
EX_DATE_FIELD_NAME=EX_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME=CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME=CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME=DIVIDEND_TYPE_FIELD_NAME,
)
_expected_fields = frozenset({
@@ -132,6 +138,7 @@ class BlazeDividendsByExDateLoader(BlazeEventsLoader):
EX_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME,
})
concrete_loader = DividendsByExDateLoader
@@ -167,6 +174,7 @@ class BlazeDividendsByPayDateLoader(BlazeEventsLoader):
{PAY_DATE_FIELD_NAME}: ?datetime,
{CASH_AMOUNT_FIELD_NAME}: ?datetime,
{CURRENCY_FIELD_NAME}: ?string,
{DIVIDEND_TYPE_FIELD_NAME}: ?string,
}}
Where each row of the table is a record including the sid to identify the
@@ -182,7 +190,8 @@ class BlazeDividendsByPayDateLoader(BlazeEventsLoader):
SID_FIELD_NAME=SID_FIELD_NAME,
PAY_DATE_FIELD_NAME=PAY_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME=CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME=CURRENCY_FIELD_NAME
CURRENCY_FIELD_NAME=CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME=DIVIDEND_TYPE_FIELD_NAME
)
_expected_fields = frozenset({
@@ -191,6 +200,7 @@ class BlazeDividendsByPayDateLoader(BlazeEventsLoader):
PAY_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME,
})
concrete_loader = DividendsByPayDateLoader
zipline/pipeline/loaders/dividends.py
+42 -3
View File
@@ -2,6 +2,7 @@ from zipline.pipeline.common import (
ANNOUNCEMENT_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME,
EX_DATE_FIELD_NAME,
PAY_DATE_FIELD_NAME,
)
@@ -17,7 +18,8 @@ from zipline.utils.memoize import lazyval
class DividendsByAnnouncementDateLoader(EventsLoader):
expected_cols = frozenset([ANNOUNCEMENT_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME])
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME])
event_date_col = ANNOUNCEMENT_FIELD_NAME
@@ -48,11 +50,19 @@ class DividendsByAnnouncementDateLoader(EventsLoader):
CURRENCY_FIELD_NAME
)
@lazyval
def previous_type_loader(self):
return self._previous_event_value_loader(
self.dataset.previous_type,
DIVIDEND_TYPE_FIELD_NAME
)
class DividendsByPayDateLoader(EventsLoader):
expected_cols = frozenset([PAY_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME])
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME])
event_date_col = PAY_DATE_FIELD_NAME
@@ -99,11 +109,26 @@ class DividendsByPayDateLoader(EventsLoader):
CURRENCY_FIELD_NAME
)
@lazyval
def previous_type_loader(self):
return self._previous_event_value_loader(
self.dataset.previous_type,
DIVIDEND_TYPE_FIELD_NAME
)
@lazyval
def next_type_loader(self):
return self._next_event_value_loader(
self.dataset.next_type,
DIVIDEND_TYPE_FIELD_NAME
)
class DividendsByExDateLoader(EventsLoader):
expected_cols = frozenset([EX_DATE_FIELD_NAME,
CASH_AMOUNT_FIELD_NAME,
CURRENCY_FIELD_NAME])
CURRENCY_FIELD_NAME,
DIVIDEND_TYPE_FIELD_NAME])
event_date_col = EX_DATE_FIELD_NAME
@@ -149,3 +174,17 @@ class DividendsByExDateLoader(EventsLoader):
self.dataset.next_currency,
CURRENCY_FIELD_NAME
)
@lazyval
def previous_type_loader(self):
return self._previous_event_value_loader(
self.dataset.previous_type,
DIVIDEND_TYPE_FIELD_NAME
)
@lazyval
def next_type_loader(self):
return self._next_event_value_loader(
self.dataset.next_type,
DIVIDEND_TYPE_FIELD_NAME
)